This work is copyright. Apart from any fair dealing for the purpose ofstudy, research, criticism or review, as permitted under the CopyrightAct, no part may be reproduced by any process without writtenpermission. Copyright is the responsibility of the Director Publishingand Marketing, AGPS. Enquiries should be directed to the Manager,AGPS Press, Australian Government Publishing Service, GPO Box 84,CANBERRA ACT 2601.
A
AR-006-602
DEPARTMENT OF DEFENCEDEFENCE SCIENCE AND TECHNOLOGY ORGANISATION
AERONAUTICAL RESEARCH LABORATORY
Aircraft Materials Technical Memorandum 404
METALLOGRAPHY OF ALUMINIDE COATINGS
by
P. ROHAN
SUMMARY
Metallographic techniques have been established for the examination of gas turbineblades and their protective coatings Particular attention was given to edge retention ofspecimens during polishing and to etchant selection to allow microstructural evaluation ofmajor phases. Edge retention was achieved firstly by depositing a nickel plate over thespecimen (by electroless deposition) and secondly, by minimizing the use of cloth polishingpads. Etchants were then evaluated for phase delineation in polished Allison T-56 turbineblades, (IN738 substrate alloy coated with high-activity nickel aluminide). Characteristics offour most useful etchants are described.
OSTOtMELBOURNE
@ COMMONWEALTH OF AUSTRALIA 1991
POSTAL ADDRESS: Director, Aeronautical Research Laboratory506 Lorimer Street, Fishermens Bend 3207Victoria Australia
CONTENTS
P!are No.
D. IN RODU T ON ........................................................................................ 1
The Corrosion Control group at ARL has commenced a programme for evaluatingprotective coatings for gas turbine blades currently used by the RAAF (workpreviously performed at MRL). To enable examination and evaluation ofmetallographic specimens (with particular attention given to surface features).techniques have been developed for sample preparation -and microstructural -etching. In developing these techniques. Allison T-56 turbine engine blades castfrom a nickel-based superalloy (IN 738) and coated with nickel aluminide weresectioned. The nickel aluminide coating applied by a high-activity "pack-cementation" process to the outer surface of the blades, Impartsoxidation/corrosion resistance to the substrate superalloy. Metallographic
procedures were developed with an aim to minimise edge rounding and highlight thedetailed microstructure by selective etching. As the coatings are only between 50and 100 pm thick and attention needs to be given to oxide/corrosion scales producedon the surface, edge retention is essential.
Nickel-based superalloys are composed of FCC y - NI solid solution containing Y'-Ni3AI precipitation phase, carbides and various other precipitates such as T1, (hcp. egNi3T), a, (tcp, eg CrCo, CrNiMol, etc.( ref. I). The coating mainly consists of 0 - NWAand may contain precipitate particlea and carbides derived from alloying elementscontained in the substrate (ref. 2). Various etchants have been used to reveal thesefeatures with the most useful being reported here.
The blades used were rejects from routine maintenance performed by the RAAF onAllison T56 turbo-prop engines (from either the C-130 Hercules or P-3 Orionaircraft) and as such showed various defects and coating conditions.
11
2.0 EDGE RETENTION
In order to minimise edge roundtng during (manual) polishing of specimens. 2 steps
were taken:
1. A NI coating is first plated over the sample,
(Ni electroplating of specimens was previously used by MRL
for related work)
2. SIC paper was used down to 4000# (511m) to minimise the use ()I cloth
pads (which exacerbate edge rounding).
2.1 NICKEL PLATING
To av'id loss of corrosion products from the outer surface of samples during
Immersion in plating solutions and due to handling, entire blades were vacuum
impregnated with epoxy resin. The viscosity of the epoxy was sufficiently low to
allow penetration of the scale and prevent fragmenting during polishing. After
impregnation, the excess resin was allowed to run off leaving only a thin film to
cure over the surface. The cured epoxy must have high hardness and minimal
shrinkage to prevent scale separation from the coating.
400# SIC paper was later used to roughen the cured epoxy surface to allow
mechanical 'keying-in' of the plated coating and avoid flaking.
The procedure for the electroless deposition of nickel is given in Table 1. Firstly,
the surface to be plated was clear I of contaminants by placing the sample in a
beaker of acetone and cleaning ultrasonically. After drying, the sample is placed in
the sensitizing-nucleating solution where the following reaction takes place (ref. 3):
Sn2+(aq) + Pd2 +(aq) Sn4+(aq) + Pd (s)
ie. palladium chloride is reduced to colloidal Pd metal particles by Sn 2 +, which then
forms hydrolysed Sn4 + which In turn stabilises the Pd colloid.
Before the sample can be plated in the electroless Ni bath, the Sn 4+ ions
surrounding the adsorbed Pd must be removed, This 'acceleration' step dissolves the
Sn4+ without affecting the Pd layer.
2J
The electroless Ni bath itself contains essentially,
a) a Ni salt (NiSO4 .6H2 0).
b) a reducing agent (sodium hypophosphite. NaH 2 PO4 .6H2 0),
Sodium acetate (NaC2 H3 02) is added to increase the deposition rate and buffer the
solution, while the other additive, lead acetate (Pb(C2 H3 02)2 2H2 0) stabilises the
bath to prevent catalytic decomposition (ref. 3).
An alternative to electroless deposition is to sputter a conducting film over the
epoxy and then electroplate a Ni coating using the low stress sulphamate bath (ref.
1): this would be useful for a small number of specimens as the electroless bath has a
short life (the procedure for electroplating is given in appendix 1).
After plating, the specimens are mounted in epoxy resin for polishing (see figs. I &
2).
2.2 POLISHIG
When the mounting epoxy had cured, the samples were abraded with SiC paper
ranging from 220# to 4000# using ethanol for cleaning and lubrication.
Interchangeable SiC paper discs were used which clipped onto a brass polishing
wheel. Care was taken to retain the brittle scale during polishing which, besides
being lost, would scratch the specimen on breaking free. After 4000#. final
polishing was carried out using 1pm diamond paste on a felt pad.
Edge rounding still occurs to some degree; however, for better results, automatic
polishing would be required.
I3
ik
3.0 ETCHANTS
Many etchants have been tried, the most useful are those listed in Table 2. Relevantfigures (with detailed captions) and microstructures are referenced in the table.
In Table 2. it may appear that the lactic acid etch (etchant (1)) and the Iron chloride
etch (etchant (2)) perform similarly; however they are quite different. Etchant (1)was used for microstructural observation of both substrate and coating. to good -effect. Etchant (2) was too severe on the aluminide coating for optical microscopy
and was more useful as a macroetchant. as it colour contrasted the dendritic
structure of the grains (fig. 8). As a consequence of its preferential attack of Al richphases, it has been used to strip the coating from a blade leaving the skeleton of the
scale, coating grain boundaries and the precipitation region of the substrate (fig. 14).
The electrolytic etchants can be used for phase extraction; ie. electrolyte (1). whenused for longer periods, can be used to dissolve the substrate leaving y' precipitates
(fig. 9) and electrolyte (2) can be used for carbide extraction (fig. 11). as little
dissolution of these phases occur, compared to the surrounding material.
3.1 CARBIDE STAIN
Figures 15 and 16 show examples of the effect obtained using the carbide stain. Two
different superalloys are shown; in figure 15, IN 738 was stained to reveal the blocky
matrix carbides of M6C type; while in figure 16, the matrix carbides are M23C6 in
Rene 80H.
A chain of small discrete carbide particles is developed in superalloy grain
boundaries during post-forming aging to improve creep rupture life (ref 1). Due tothe presence of these carbides in grain boundaries, the carbide stain can also be used
to delineate grains on a polished specimen without destructive etching, (fig. 17).
4
4.0 CONCLUSION
1, Edge retention of polished turbine blade specimens is essential for
examination of the aluminide coating applied to the outer surface. This has
been acheived by nickel plating an approximately 30 irn thick layer over the
specimen before mounting and polishing.
2. Four etchants have been selected for their ease of use and phase delineation:
I a microetchant (immersion) for coating and substrate.
ii. a macroetchant (immersion) for the substrate.
iii. an electrolytic microetchant for the substrate and for phase
extraction of y' .
iv. an electrolytic microetchant for substrate and coating, and
pbsae extraction of carbides.
In addition, a carbide stain was used to identify carbide phases by colour
contrasting.
i
5.0 RZFERMlg CES
1. ASM. Metals Handbook. 9th edition.
2. G. R. Johnston. P. G. Richards. "High-Temperature Corrosion in Military Gas
Turbines: A Metallographic Comparison of ALPAK Aluminide and CAC
Alurninide Coatings for First-Stage Turbine Blades In RAAF T-56 Engines".
MRL report. MRL-R-817. (1981).
3. F. A. Lowenhelin. "Electroplating'. McGraw-Hill Pub. (1978).
4. R Kossowsky. S. Singhal (eds.). "Surface Engineering". Martinus NiJhoff
Pub.. (1984).
5. ASTM. "Standard Methods For Microetching Of Metals and Alloys".
Designation E407. Etchant No. 40.
6. E. Kny. "On The Methodology of Phase Extraction in NI Based
7. P. J. Fink, R, W. Heakel. "Analysis of Microstructural Change due to CyclicOxidation in Aluminilde-coated NI-Al. NI-Cr and NI-Cr-Al Alloys". HighTemperature Coatings. The Metallurgical Sovc!'ly Pub.. (1986).
6
Instead of the electroless plating procedure. a Ni cnating may be applied by
electroplating if a conducting film exits on the surface. A conducting film may be
applied simply by sputtering techniques; however, of the metals tried, gold
performed the best so this was used sparingly.
The electroplating bath that was most satisfactory, by way of solution stability and
producing a coating of low stress, was the Sulphamate Bath (ref. 1):
4. Electrolyte (2)(ref. 7)10% HCI 2V. 10 sec* Dissolves coating &4% Tartaric acid substrate leaving carbidesin Methanol and interdiffusion zone
precipitates.(figs. 11 to 13)
Carbide stain (ref. 5): Swab 3 sec Stains carbides according to50/50 Aqueous soin. structure:K2MnO 4 and 60/o - M23C6 green-pale blueNaOH soln. - M6 C brown
- MC multicoloured.
(figs. 15 to 17)
* For a specimen area of approximately 1.5cm2 .
9
Im m ......... m d I IM ~ ilIl I I mlmm
1: Edge and corrosion product retention using theNi plating procedure.
Eigm2 Retention of corrosion product produced bystatic laboratory tests; the bright band on LHS is the bladematerial, RHS is NI plate.
1
EW 3 SEM micograph of the effect of etchant (1) on a highactivity aluminide coating. On the RHS of the micrograph. the -matrix contains voids where y' has been removed. To the left is theinterdiffusion zones which are highlighted due to dissolution of '(inner) and p (outer) phases which previously surrounded theprecipitates. The coating 1 phase has been dissolved revealingprecipitates arising from substrate alloying elements in two distinctbands; a finer precipitate size close to the outer interdiffusion zone.and a coarser precipitate toward the outer edge. Little oxidation ofthis coating has occurred as there is no indication of -' formation dueto Al depletion, (ref.7).
L, ' 11
Eg= 4: Optical micrograph showing the effect of etchant(1) on the substrate and coating. The coating grain structure isrevealed (showing less precipitates than fig. 3 due to solutioningduring service) and the substrate -t, y', carbides and grain boundariesare clearly seen.
4L
END= 5: Also etchant (1), but here the coating has undergoneAl depletion as the A phase exists only as isolated grains in thecoating, the rest of the coating is probably y' phase.
12
Fig= 6:~ SEM micrograph obtained using etchant (2).This etchant has revealed more coating structure for a similarcoating area to fig. 4 due to preferential attack on Al rich regions.Because of this, etchant (2) is not as applicable for opticalmicrographs as etchant (1).
Eli= 7: A different coating area than in fig. 6 (also etchant (2))showing greater solutioning of coating precipitates due to longerservice exposure. Oxidation is also visible down a coating grainboundary reaching the substrate.
13
igure 8: Colour contrasting effect of etchant (2) onmacrostructure. The dendritic grain structure and corrosiondepleted regions are clearly defined (LHS is leading edge).
EWgu9: In this micrograph, etchant (3) has been used to isolatey' in the substrate (LHS). The p phase of the coating is intact and theinterdiffusion zone precipitates have dissolved.
14
I,!
Eigm 10 Optical micrograph using etchant (3). The darker phaseat the outer edge of the coating is probably y' formed by Al depletion.Grain-boundary carbides are also visible in the substrate.
11 Etchant (4); p3, y and y'phases have been preferentiallyattacked leaving carbide phases and interdiffusion zone precipitates.Note the plate-like appearance of precipitates projecting into thesubstrate (RHS).
15
Mum 12: Etchant (4) used to reveal grain boundary carbides.y' attack has left voids in the matrix and grain boundaries.
igm1: Optical micrograph showing coating and substrateetched with etchant (4) for 10 sec.
16
Fgr 14: SEM micrographs of the outer surface of a turbine bladewith NIAl removed by etchant 2;
a) shows the scale (top 1/2), and the interdiffusion zone, (bottom1/2), connected by coating grain boundaries.
b) magnification of grain boundaries in a),
C) three regions of the coating are revealed here, starting fromthe LHS bottom,
-inner and outer interdiffusion zone bands-Ni rich region of the coating (smoother diagonalband).
17
Il
18
Ejgig15: Optical micrograph showing the effects of carbide stainon superalloy 1N738. Larger matrix carbides and smaller chain grainboundary carbides have stained brown (M6C).
flgurgM M23 C6 carbides dispersed in superalloy Rent 80H.
19
EkguzIA7i Carbide stain used to delineate structure without etching.The micrograph reveals a fatigue crack running down a substrategrain boundary.
20
DISTRIBUTION
AUSTRALIA
Department of Defence
Chief Defence ScientistAS. Science Corporate Management ) shared copyFAS Science PolicyDirector, Departmental PublicationsCounsellor, Defence Science, London (Doc Data Sheet Only)Counsellor, Defence Science, Washington (Doc Data Sheet Only)Scientific Adviser, Defence CentralOIC TRS, Defence Central LibraryDocument Exchange Centre, DSTIC (8 copies)Defence Intelligence OrganisationLibrarian H Block, Victoria Barracks, MelbourneS.A. to Thailand MRD (Doc Data Sheet Only)S.A. to the DRC (Kuala Lumpur) (Doc Data Sheet Only)Dircctor General - Army Development (NSO) (Doc Data Sheet Only)Defence Industry and Materiel Policy, FAS (Doc Data Sheet Only)
Aeronautical Research LaboratoryDirectorLibraryChief, Aircraft Materials DivisionDivisional File, Aircraft MaterialsAuthor: P. RohanN.E. Ryan
Materials Research LaboratocyDirector/Library
Navy OfficeNavy Scientific Adviser (3 copies Doc Data Sheet)Director of Aviation Systems Engineering - NavyDirector of Naval Engineering Requirements -Aviation SystemsNaval Support Command
Superintendent, Naval Aircraft Logistics
AmuX fflcScientific Adviser - Army (T)oc Data Sheet only)
AirFe OfficeAir Force Scientific AdviserDirector General Engineering - Air ForceAHQ (INT I)
Statutory and State Authorities and IndustryAero-Space Technologies Australia, Systems Division Librarian
Ansett Airlines of Australia, LibraryAustralian Airlines, LibraryQantas Airways LimitedCivil Aviation AuthorityHawker de Havilland Aust Pty Ltd, Victoria, Library
Hawker de Havilland Aust Pty Ltd, Bankstown, Library
Rolls Royce of Australia Pty Ltd, Manager
BHP, Melbourne Research Laboratories
Universities and CollegesAdelaide
Barr Smith Library
MelbourneEngineering Library
MonashHargrave LibraryHead, Materials Engineering
NewcastleLibraryProfessor R. Telfer, Institute of Aviation
SydneyEngineering Library
NSWPhysical Sciences LibraryLibrary, Australian Defence Force Academy
Metalloartuhic techniques have been established for the examination of ;as turbine ba ad theirprotective coatings. Particular attention was gawn to edge retention of spectmens durin polishing and toetchw selection to allow ,paostructural evltion of majr phases. Edge retention ums achiredfirstlyby depositing a Nickel plate ower the specimn (by eectroles deposition) and secondiy wuhntheuse of cloth polishing pads. Etchants were then euzluated for phase delineation In =Affied.AV5hs
urbne laes,(P138substrate alloy coated with hlgh-acuWry nickel almmunide). Characteristic of fourmvs usefla etchawt are describedl
-AGI aIAB~cAXIl
UNCLASSIFIED
ff13 PAGE ism 10B UW)70go = I1VarW W1U4 IF S aQUWW MI71 WrANJGHUf, FU IM OW EIU00 Wuvfl HIVTBE AZW T0 IIHE uDIBM AL aUUL9 .'SaCALY fQUZMMh
WS ABUI1RPCr MOMf
17. 11TM
AERONAUICAL RESEARCH LABORATORY. MLDOURNE
1S. DOCUENr CIU AND NUBI to. cour 20 WlVgO~tlW~AND W OD
